Set the investigation boundary
Identify the affected service and the question the agent should investigate using its available observability access.
Sense / AI operations
Give agent-assisted investigation operational context.
Sense connects agent-assisted investigation to observability data through AI SRE and MCP within Unified Observability. A scoped investigation can use service telemetry to develop a useful engineering handoff while keeping the evidence behind its conclusions available for review.
AI SRE & MCPSense
Investigate the invoice export incident using the available observability context and prepare findings for the on-call engineer.
In the workspace
Investigation scope
Invoice export incident · Read context
Data connection
MCP · Available service telemetry
Engineer handoff
Evidence and unresolved questions
Connect the investigation agent to the service evidence it needs.
How it works in Sense
The on-call engineer receives an investigation summary that distinguishes observed behavior from a suspected explanation. The handoff identifies what still needs checking so the engineer can review the evidence before deciding on an operational response.
Gather relevant operational context
Use the connected telemetry to examine the failing request path and identify evidence related to the incident.
Prepare a reviewable handoff
Summarize the findings with their supporting context and state which assumptions still require engineering validation.
Bring Sense to your team.
Explore ai sre & mcp around your company’s workflows.